Role of preconception nutrition supplements in maternal anemia and intrauterine growth: a systematic review and meta-analysis of randomized controlled trials

Table 2 Summary of findings for the quality of evidence for maternal hemoglobin, birth weight, birth length, birth head circumference, small for gestational age, and preterm birth using a GRADE assessment tool

Outcomes	Number of RCTs (study participants)	Pooled summary measure^a and 95% CIs	I² (%)	Final rating for quality
Maternal hemoglobin (g/dL)	9 (5,630)	0.30 g/dL (0.03, 0.57)	79	Moderate
Birth weight (gm)	10 (12,929)^c	12.25 gm (− 22.66, 47.16)	55	Very low
Birth length (cm)	5 (4,212)	0.15 cm (− 0.26, 0.56)	68	Very low
Birth head circumference (cm)	4 (4,178)	− 0.23 cm (− 0.88, 0.43)	84	Very low
Small for gestational age	8 (13,856)	RR 0.91 (0.80, 1.04)	31	Moderate
Preterm birth	12 (18,413)	RR 0.93 (0.69 1.25)	57	Very low

RR risk ratios, 95% CIs 95% Confidence intervals
^aPooled summary measures are mean differences for outcomes on a continuous scale including maternal hemoglobin (g/dL), birth weight (gm), birth length (cm), and head circumference (cm) and risk ratios (RR) for dichotomized outcomes (small for gestational age and preterm birth)
^bFive GRADE criteria (Risk of bias, inconsistency, imprecision, indirectness, and publication bias) were used to assess the quality of evidence (see Supplementary Table S4 for details)
^cOverall, 15 RCTs measured birth weight; however, only 10 RCTs provided sufficient raw data to compute the pooled mean difference in birth weight (gm) between intervention and control Arms
The four circles with the red color represent the rating for the quality of evidence. High quality of evidence: all four circles were made red; Moderate quality: three circles were colored red; Low quality of evidence: two circles were made red; Very low quality of evidence: only one circle was made red
Very low: the calculated effect size is very different from the truth
Low: the measured effect size may be substantially different from the truth
Moderate: The calculated effect measure may be close to the truth, but one may find a substantial difference between the estimated mean difference and the truth
High: there is a high confidence that calculated estimate is very close to the truth, and further studies may not find a substantial difference

ISSN: 2046-4053